The proposed project is concerned with one of the most basic and fundamental forms of learning, Pavlovian conditioning. This type of learning involves coming to anticipate a biologically significant event or unconditioned stimulus (US) on the basis of an earlier signal or conditioned stimulus (CS). Prior research has employed arbitrary conditioned stimuli that have no inherent relation to the US, and the work has focused on how new responses come to be elicited by such conditioned stimuli. The proposed research is motivated by a new functional perspective on Pavlovian conditioning. Functional considerations suggest that Pavlovian learning in natural environments involves conditioned stimuli that have some pre-existing relation to the US, and the important consequence of the learning is not CS-elicited behavior but increased effectiveness in how the organism responds to the unconditioned stimulus. The proposed studies will examine this new perspective in studies of Pavlovian conditioning in the sexual behavior system, using nonhuman animal subjects: One set of experiments will examine learning with conditioned stimuli that are not arbitrary but include limited species typical cues. Another set of experiments will examine how Pavlovian conditioning alters unconditioned sexual behavior and produces changes in reproductive success. The research will further our understanding of Pavlovian learning and will integrate the functional approach of ethologists with the emphasis on learning mechanisms in traditional studies of associative learning.